Unverified50% confidenceFactExact time
Figure's robots and Tesla's Optimus are being tested on tasks such as sorting and simple assembly, using vision-language-action models to map camera images and voice commands into action sequences.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/13/2026
First Seen
Valid until: 1/11/2027
Sources
Related Entities
Related Claims
VerifiedVLA directly connects pre-trained vision-language models with robot action policies, enabling robots to process natural language instructions, camera images, and low-level motion control in a unified manner.73% similarUnverifiedUsing representations from vision-language models such as CLIP, few-shot performance on robot grasping tasks can be improved by 30–50%72% similarUnverifiedThe project is a pose-based teleoperation system that captures human body pose in real time via an ordinary webcam and maps it onto the joints of a quadruped robot.71% similarUnverifiedThe task of automated screen recording tools reproducing camera movement decisions is fundamentally an intent recognition problem, as the same click in a navigation jump versus a form-filling scenario may or may not warrant a zoom.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1022408API
curl https://kongchang.com/api/v1/knowledge/claims/1022408MCP
get_claim(id=1022408)